ECS-9701539 Gorur Power system reliability is dependent on efficient and accurate characterization of network components. This proposal addresses the critical need for condition assessment and early failure detection in polymeric outdoor equipment, key elements in the power delivery system. The present use of polymeric insulation in power systems is estimated at US $ 2 billion per year and increasing, with USA being the single largest user. Flashover, or the inability of the device to maintain the service voltage, and housing material degradation, which can lead to disastrous consequences such as dropping of the line, are the principles failure modes experienced in service. Until now, there is no reliable method available to predict when failure of the insulation, which is one of the weakest link in the system, will occur. Our interdisciplinary team of insulation systems expertise and statistical analysis expertise proposes to focus on making significant breakthroughs to enable condition assessment and early failure detection by demonstrating that: (1) It is possible to quantify electrical discharge activity that initiates material degradation, (2) Statistically efficient designed experiments and models can be used to create a predictive model linking device characteristics and service conditions to material degradation and flashover, and (3) A computer model which incorporates detailed device design can reliably predict flashover occurrence, and consequently can be used for preventive maintenance in the system. Advancements in the understanding of aging phenomena of polymers, electric field calculation, flashover mechanism, data acquisition, signal processing related to electrical discharges, material characterization and high speed photography will be utilized in this project. The failure criteria will be translated into parameters that are easy to use and in industry standard terms. Failures will be detected from measurements of discharge current, arc dynamics, discharge tempera ture and chemical detection process. Each of these criteria has potential for development as a practical sensor device that can be used for preventive and predictive maintenance.
